We are conducting a positional cloning project to identity a putative tumor suppressor gene(s) on chromosome 3p12.2 or 3q25.3, which are regions previously identified by allelic deletion and cytogenetic analysis, to harbor candidate tumor suppressor genes. Koji Sasajima, a former LHC fellow, has identified a colorectal cancer patient with diffuse digestive tract polyposis and germ line abnormality in chromosome 3; inv(3)(p12.2q25.3). Therefore, we proposed the hypothesis that the gene(s) affected by the chromosome inversion predisposed the patient to diffuse polyposis and malignancy. The chromosomal 3 breakpoints of this patient have been isolated by positional cloning strategy. We and our collaborator microscopically dissected the chromosome fragments around the breakpoints. Using a DNA probe isolated from the dissected fragments, we made YAC contigs and YAC clones containing the breakpoint in 3p which were identified by FISH (fluorescent in situ hybridization). The cosmid contigs were made from these YACs, cosmid clones spanning the breakpoint were identified by FISH, and the nucleotide sequence at and around the breakpoint in both 3p12.2 and 3q25.3 were determined. The patient was found to have 2bp deletion in 3p12.2 and about 300kb deletion in 3q25.3. An extensive search for the gene(s) disrupted by the chromosome inversion is in progress by two ways; (1) determination of the nucleotide sequence around the breakpoints - 120 kb in 3p and 130 kb in 3q have been determined and (2) long-range exon trapping which is a recently developed procedure. The candidate transcribed sequences are to be studied by investigating their abnormalities in cancer cells. We have so far obtained two clusters possible transcribed sequences, both of which have significant homology with the EST(expressed sequence tag) clones of unknown function.